Title:	Prompt engineering and prompt chaining in artificial intelligence: tools for mapping future climate scenarios as mechanisms of adaptation and climate justice and just transition promotion.
Authors:	FONTENELLE, M. R. AMORIM, J. R. A. de CRUZ, M. A. S. LIMA, C. E. P.
Affiliation:	MARIANA RODRIGUES FONTENELLE, CNPH; JULIO ROBERTO ARAUJO DE AMORIM, CPATC; MARCOS AURÉLIO SOARES CRUZ; CARLOS EDUARDO PACHECO LIMA, CNPH.
Date Issued:	2025
Citation:	Revista DCS, v. 22, n. 83, p. 1-17, 2025.
Description:	The climate emergency demands innovative methodological approaches to reach socioeconomically and environmentally vulnerable populations and areas as quickly as possible. This article proposes the use of Open Science, Prompt Engineering, and Prompt Chaining as tools for mapping future climate scenarios using automated methods produced with the aid of Generative Artificial Intelligence (AI – LLMs models). Based on an applied study, it demonstrates how these techniques can be integrated into climate mapping models focused on family farming, in line with IPCC frameworks on climate justice and just transition, significantly increasing productivity, scalability and reducing time and potential errors. The results indicate that AI, when guided by well-structured prompts and chained logical flows, can democratize access to complex analyses, support public policies, and strengthen the resilience of vulnerable communities. The workflow proposed here can be replicated for other areas, models, climate scenarios, and agricultural crops. The deposit of all Python scripts generated on the Zenodo open access platform is in line with the FAIR principles of open science
Thesagro:	Mudança Climática Agricultura Familiar
Keywords:	Objetivos de Desenvolvimento Sustentável (ODS) Engenharia de Prompt Inteligência Artificial Generativa Modelos de Linguagem de Grande Escala (LLMs) Mapeamento Climático Automatizado Cenários Climáticos Futuros Ciência Aberta Geoprocessamento automatizado
ISSN:	2224-4131
DOI:	10.54899/dcs.v22i83.3456
Type of Material:	Artigo de periódico
Access:	openAccess
Appears in Collections:	Artigo em periódico indexado (CNPH)